Sei sulla pagina 1di 18

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/262783591

Worldwide incidence of cervical lesions: A systematic review

Article  in  Epidemiology and Infection · May 2014


DOI: 10.1017/S0950268814001356 · Source: PubMed

CITATIONS READS

8 288

7 authors, including:

Jie Ting Anne Fortino Rositch


University of California, San Francisco Johns Hopkins Bloomberg School of Public Health
27 PUBLICATIONS   268 CITATIONS    70 PUBLICATIONS   1,002 CITATIONS   

SEE PROFILE SEE PROFILE

Sylvia Taylor Lisa Rahangdale


GlaxoSmithKline University of North Carolina at Chapel Hill
36 PUBLICATIONS   762 CITATIONS    55 PUBLICATIONS   349 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Human papillomavirus View project

Time to A.C.T.: Implementing strategies for breast cancer control in Africa View project

All content following this page was uploaded by Heidi M Soeters on 07 August 2014.

The user has requested enhancement of the downloaded file.


Epidemiol. Infect., Page 1 of 17. © Cambridge University Press 2014
doi:10.1017/S0950268814001356

REVIEW ARTICLE
Worldwide incidence of cervical lesions: a systematic review

J. TING 1 *, A. F. ROSITCH 2 , 3 , S. M. TAYLOR 4 , L. RAHANGDALE 5 ,


H. M. SOETERS 6 , X. SUN 6 A N D J. S. SMITH 6 , 7
1
Department of Clinical Pharmacy, University of California, San Francisco, CA, USA
2
Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
3
Department of Epidemiology and Public Health, University of Maryland School of Medicine, Baltimore,
MD, USA
4
GlaxoSmithKline Vaccines, Global Vaccine Development, Wavre, Belgium
5
Department of Obstetrics & Gynecology, University of North Carolina at Chapel Hill, NC, USA
6
Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel
Hill, NC, USA
7
Lineberger Comprehensive Cancer Center, Chapel Hill, NC, USA

Received 16 December 2013; Final revision 13 April 2014; Accepted 8 May 2014

SUMMARY
We conducted a systematic review summarizing data on incidence of high- and low-grade
lesions in women with normal baseline cervical cytology, stratified by age (<30 and 530 years),
and baseline human papillomavirus (HPV) infection. Incidence of high- and low-grade lesions
in women aged 530 years with a baseline HPV infection increased over follow-up time
(5–127 months), although incidence generally remained <10%. Without baseline HPV infection,
incidence of high-grade lesions remained low over follow-up time (<5% over 5–122 months).
Incidence of high-grade lesions in women aged 530 years with baseline HPV infection appeared
similar to that in women aged <30 years. In some women aged <30 years, high-grade lesions can
develop relatively shortly after initial HPV infection. We observed an increase in low-grade
lesions over time in women aged 530 years with baseline HPV infection, potentially indicative
of an HPV infection that is potentially progressing to higher grade lesions.

Key words: Epidemiology, human papilloma virus (HPV).

I N T RO D U C T I O N intraepithelial neoplasia 2 and 3 (CIN 2/3)] [2].


Infection with high-risk human papillomavirus Persistent HR-HPV infection is considered necessary
(HR-HPV) types is the major aetiological cause of to promote progression of pre-malignant stages to in-
vasive cancer [3]. Compared to cervical cytology, cer-
cervical cancer and its pre-cancerous lesions [1], with
types 16 and 18 accounting for ∼70% of cervical vical cancer screening by molecular HR-HPV testing
cancers and ∼50% of high-grade lesions [cervical is more sensitive, although less specific for the detec-
tion of CIN 2/3 [4].
Currently in the USA, HR-HPV testing is rec-
ommended and frequently used for co-testing with
* Author for correspondence: J. Ting, PhD, MSPH, Department of cytology for screening of women aged 530 years
Clinical Pharmacy, University of California, San Francisco, 3333
California St, Laurel Height Ste 420, CA 94143, USA.
(hereafter termed older women) [5, 6]. Current guide-
(Email: tingj@pharmacy.ucsf.edu) lines do not recommend HR-HPV DNA testing in
2 J. Ting and others

women aged <30 years (younger women), since HPV were imposed. The following search terms were
infection is relatively common in this age group and included: genital lesions, cervical cancer, cervical dys-
is a less specific predictor of high-grade cervical pre- plasia, cervical lesions, squamous intraepithelial
cancer or cancer [7]. Examining the acquisition of lesions (SIL), CIN, atypical squamous cells of unde-
cervical lesions in women worldwide, stratified by termined significance (ASCUS), cohort study, and
age and by baseline HPV testing results (overall or incidence. Only studies conducted in 1989 or later
high-risk HPV), allows for a comparison of the differ- [after the implementation of the Bethesda System
ence in risk of the development of cervical lesions in (TBS)] were considered in order to reduce the possi-
older compared to younger women. bility of misclassification of cytological diagnoses.
Although HPV vaccines are also licensed for use in To be eligible, a study must have reported at least
women aged >26 years in many countries, the main population age and one estimate of the incidence of
target population for HPV vaccination is adolescent cytological or histological cervical abnormalities.
girls before the onset of sexual activity. In women Studies of HIV-seropositive populations, those with
aged 24–45 years, the quadrivalent HPV-16/-18/ baseline prevalence of cervical abnormalities exceed-
-6/-11 vaccine (Gardasil®, Merck & Co. Inc., USA) ing 15%, and those of women attending sexually
was efficacious in protecting against HPV-related transmitted diseases clinics were omitted as these
outcomes in women not previously exposed to the women were considered at a relatively higher risk of
HPV vaccine-types, but had limited protection in incident cervical lesions than the general population.
women who previously had a vaccine-type infection For HPV vaccine trials, only women in the control
[8]. Initial results of the bivalent HPV-16/-18 AS04- arm (unvaccinated population) were included in our
adjuvanted vaccine (Cervarix®, GlaxoSmithKline analyses. All search results were independently
Vaccines, UK) trial in women aged 526 years also reviewed (J.T. and B.W.) to ensure that no relevant
showed high vaccine efficacy against HPV-16/ articles were omitted.
-18-associated CIN 1+ and/or persistent infection, but
vaccine efficacy decreased when taking into account
women with previous HPV infection or disease [9]. Data abstraction
Full results of the bivalent vaccine trial will be pub- Abstracted data included (i) incidence proportion
lished shortly. However, as older women are still at (equivalent to cumulative incidence) with associated
risk of HPV infection and subsequent development standard errors or confidence intervals (CIs), and (ii)
of disease, examining patterns of cervical lesion de- time interval over which incidence was measured.
velopment in women aged >26 years can provide Study information abstracted included first author,
valuable data on the burden of disease for the future publication journal and date, country and city of the
development of cervical cancer prevention strategies study, period in which study was conducted, char-
for this age group. acteristics of study population (e.g. attendees of or-
Despite a relatively large number of studies which ganized or routine cytological screening), mean,
examined the incidence of cervical lesions stratified median or age range of study population, study ex-
by HPV infection status and by age, no systematic re- clusion criteria (e.g. current pregnancy, past hyster-
view of the literature has been conducted to date. ectomy), sample size (number of women screened),
Thus, we present here a systematic review summariz- baseline cervical status, cervical cancer diagnostic
ing data on the incidence of cervical lesions in method used at baseline, baseline HPV DNA or sero-
women with normal cervical cytology at baseline, stra- logy prevalence or type, HPV testing method [poly-
tified by population, age, follow-up time and baseline merase chain reaction (PCR) or hybrid capture 2
HPV infection status. (HC2)], PCR primers if available (MY09/11, GP5
+/6+, SPF-10, etc.), mean/median time of follow-up,
and final diagnosis of cytological [ASCUS, low-grade
METHODS squamous intraepithelial lesion (LSIL), or high-grade
squamous intraepithelial lesion (HSIL)] and/or histo-
Identification of eligible studies logical (CIN 1, 2 or 3) lesions. Abstracted data were
Published studies were identified through PubMed double-checked by two independent investigators
(1 January 1989 to 30 September 2012) and the refer- (H.S. and X.S.) and any discrepancies were resolved
ence lists of eligible articles. No language restrictions by consensus.
Cervical cancer incidence global review 3

Selection of incidence estimates Characteristics of abnormal cytology incidence and


Some articles that met the inclusion criteria were HPV testing
based on the same study population. To maintain The majority (57%) of studies reported data on inci-
the independence of study results, we chose the article dence proportion of CIN 2+ in women with baseline
with the greatest number of women in the incidence normal cytology. There are comparatively fewer
proportion analysis for cervical lesions, or the most studies reporting incidence proportion of CIN 1
recent study results, if study sample sizes did not from baseline normal cytology (18%). In terms of
vary. If multiple articles from the same study popu- data on baseline HPV testing, approximately one-fifth
lation could contribute to separate analyses (e.g. if (∼21%) of the studies included only women who had
one article provided incidence data by age, while an HPV infection, and another one-fifth (∼19%) did
another provided data by baseline HPV infection), not perform or report HPV testing. Baseline HPV
all relevant articles were included. The time at which serology was performed or reported in only 15% of
incidence proportion was measured differed across included studies.
studies, so the corresponding length of follow-up
was abstracted. For studies in which the estimated Cumulative incidence of HSIL/CIN 2/3
standard error for the incidence proportion was not
reported, it was calculated as the square root of Estimates of incidence proportion for HSIL and CIN
(p*(1 – p))/n, where p is the observed proportion of 2/3 (Fig. 1a) were stratified by baseline HPV infection
incident cases and n the sample size. (HPV positive, HPV negative, or mixed (for studies
that presented incidence proportion overall, without
stratification by HPV infection), and further stratified
Descriptive analyses
by age group (<30 years, 530 years). There was a
If an article reported multiple results that fell into greater number of studies with incidence proportion
different categories of study characteristics, the article data for HSIL/CIN 2/3 in women aged 530 years
was included in all relevant categories in Table 1 so than for women aged <30 years.
that the proportion of study results could add to In women aged 530 years who were HPV positive
more than 100%. Selected characteristics as well as at baseline, incidence proportion estimates for HSIL/
the incidence proportions and their associated 95% CIN 2/3 generally increased with longer follow-up
CI for every study included in this review are listed time (Fig. 1a). Across studies, incidence proportion
in Tables 2a and 2b, and arranged according to region estimates varied at any given follow-up time.
and by follow-up time. Figures of incidence pro- Nevertheless, across follow-up time (5–122 months),
portion over time by age and baseline HPV infection incidence proportion estimates were generally <10%,
are presented in Figures 1 and 2, and were generated except in population-based screening in The
using GraphPad Prism version 6.00 (GraphPad Netherlands (mean age 38·5 years, incidence pro-
Software, USA). In addition, we also conducted portion of CIN 2/3 at 18 months = 12·0%) [10], and
meta-regression of incidence proportion of HSIL/ in Sweden (mean age 35·1 years, incidence proportion
CIN 2/3 in women with normal baseline cytology, of CIN 2/3 at 48 months = 27·5%) [11], as well as for
stratified by baseline HPV infection (see Appendix). women in the pathology database of a hospital in
Amsterdam, The Netherlands (median age 37 years;
R E SU LTS incidence proportion of CIN 2/3 at 34 months =
18·2%) [12] (Table 2a). In women aged 530 years
Eligible studies who were HPV negative at baseline (Fig. 1b) or in
Our systematic search of the published literature iden- mixed populations, incidence proportion estimates
tified 54 studies that met the study inclusion criteria for HSIL/CIN 2/3 were generally <5%, remaining
and reported non-duplicate results of the incidence relatively constant across follow-up time (5–122
of cervical abnormalities. Briefly, approximately months) (Fig. 1c), except in population-based screen-
half (52%) of the studies were conducted in Europe ing in the UK (median age 37·5 years; mixed popu-
and 20% of the studies were conducted in North lation, incidence proportion of CIN 3 at 60 months =
America (Table 1). Most of the studies were in po- 9·0%) [13].
pulations of women with a mean or median age of Compared to women aged 530 years, there were
530 years (71%). fewer data on the incidence of cervical lesions for
4 J. Ting and others

Table 1. Characteristics of studies on the incidence of cervical lesions

No. of
results % References

Study region
Europe 28 52 [10–15, 23, 29–49]
North America 11 20 [17, 50–59]
Central and South America 5 9 [26, 60–63]
Asia including Australia* 3 6 [64–66]
Multi-region 7 13 [16, 22, 67–71]
Study populations
Population-based† 20 37 [10, 11, 13–15, 23, 26, 35–41, 43, 47, 48, 64–66]
Screening-based 17 31 [30–33, 42, 44, 45, 49, 51–53, 55, 58, 59, 61–63]
GYN/family-planning/hospital-based 4 8 [12, 29, 50, 57]
Young women (mean age 18–21 years) 13 24 [16, 17, 22, 34, 46, 54, 56, 60, 67–71]
Mean/median age of women (years)‡
<20 7 13 [16, 17, 46, 53, 54, 60, 67]
20–24 7 13 [22, 34, 56, 68–71]
25–29 4 8 [15, 23, 57, 59]
30–34 7 13 [42, 45, 58, 61–63, 65]
35–39 16 30 [10–14, 29, 31–33, 35, 37, 45, 51, 52, 64, 66]
540 15 28 [15, 26, 30, 36, 38–41, 43, 44, 47, 48, 50, 55]
HPV DNA detection method
MY09/11 9 16 [13, 17, 26, 53, 54, 57, 58, 62, 63]
GP5+/GP6+ 11 20 [10–12, 23, 36–38, 40, 45–47]
SPF10-LiPA 4 8 [16, 22, 70, 71]
Hybrid capture 2 (HC2) 14 26 [15, 29–32, 34, 42, 44, 48, 50–52, 61, 66]
Mixed§ 3 6 [14, 33, 49]
Not specified 5 9 [56, 59, 67–69]
No HPV detection 8 15 [35, 39, 41, 43, 55, 60, 64, 65]
Baseline cervical status
Normal 50 92 [10–17, 23, 26, 29–66, 70, 71]
Normal/ASCUS/LSIL/HSIL 4 8 [22, 67–69]
Definition of incidence
Normal to ASCUS 11 20 [15, 23, 43–46, 50, 55, 57, 60, 61]
Normal to ASCUS/LSIL 4 8 [30, 31, 34, 63]
Normal to ASCUS+ 7 13 [17, 29, 38, 42, 52, 65, 70]
Normal to LSIL+ 18 33 [14, 15, 17, 23, 43, 45, 46, 50, 52–57, 60–62, 70]
Normal to HSIL 16 30 [23, 30, 31, 33, 34, 43, 45, 46, 50, 52, 54, 55, 57,
60, 61, 63]
Normal to CIN 1 10 18 [12, 16, 32, 34, 39, 40, 44, 61, 64, 70]
Normal to CIN 2/3 31 57 [10–13, 16, 17, 26, 29–32, 34–41, 44, 46–49, 51,
58, 59, 61, 64, 66, 71]
Normal/ASCUS/LSIL/HSIL without HPV-16 2 4 [22, 69]
to HPV-16-associated CIN 1
Normal/ASCUS/LSIL/HSIL without HPV-18 2 4 [22, 69]
to HPV-18-associated CIN 1
Normal/ASCUS/LSIL/HSIL without HPV-6 2 4 [22, 69]
to HPV-6-associated CIN 1
Normal/ASCUS/LSIL/HSIL without HPV-11 2 4 [22, 69]
to HPV-11-associated CIN 1
Normal/ASCUS/LSIL/HSIL without HPV-16 4 8 [22, 67–69]
to HPV-16-associated CIN 2/3
Normal/ASCUS/LSIL/HSIL without HPV-18 4 8 [22, 67–69]
to HPV-18-associated CIN 2/3
Normal/ASCUS/LSIL/HSIL without HPV-6 4 8 [22, 67–69]
to HPV-6-associated CIN 2/3
Cervical cancer incidence global review 5

Table 1 (cont.)

No. of
results % References

Normal/ASCUS/LSIL/HSIL without HPV-11 4 8 [22, 67–69]


to HPV-11-associated CIN 2/3
Mean/median length of follow-up
424 months 12 22 [10, 23, 29, 32, 33, 37, 41, 42, 50, 55, 61, 65]
25 months 42 78 [11–17, 22, 26, 30, 31, 34–36, 38–40, 43–49,
51–54, 56–60, 62–64, 66–71]
Baseline HPV status‡¶
0% positive 12 22 [14, 16, 17, 30, 45, 46, 49, 58, 59, 69–71]
1–10% positive 10 18 [11, 13, 15, 34, 36, 38, 40, 44, 47, 50]
11–24% positive 10 18 [15, 22, 31, 48, 51, 62, 63, 66–68]
25–49% positive 5 9 [23, 29, 54, 56, 57]
50–74% positive 2 4 [32, 42]
75–100% positive 12 22 [10, 12, 14, 17, 26, 30, 37, 45, 52, 53, 59, 61]
No HPV detection 9 16 [33, 35, 39, 41, 43, 55, 60, 64, 65]
Baseline HPV serostatus
HPV-16/-18/-6/-11 seronegative 3 6 [16, 22, 69]
HPV-16/-18 seronegative 5 9 [54, 67, 68, 70, 71]
No HPV antibody detection 46 85 [10–15, 17, 23, 26, 29–53, 55–66]

ASCUS, Atypical squamous cells of undetermined significance; CIN, cervical intraepithelial neoplasia; GYN, gynecological
clinic; HPV, human papillomavirus; HSIL, high-grade squamous intraepithelial lesion; LSIL, low-grade squamous intrae-
pithelial lesion; NBCCEDP, National Breast and Cervical Cancer Early Detection Program.
* Includes two studies from Australia [64, 65].
† Population-based studies from the following regions: Denmark [15, 23]; Italy [35]; The Netherlands [10, 36–41, 47]; Norway
[11]; Sweden [14]; UK [13, 43, 48]; Costa Rica [26]; Australia [64, 65] and China [66].
‡ Kjaer et al. [15] included a younger population (22–32 years) and an older population (40–50 years); Cushieri et al. [45]
included a population with baseline high-risk (HR)-HPV infection (median age 30 years) and a population without baseline
HR-HPV infection (median age 38 years)
§ HPV DNA detection by HC2 followed by GP5+/6+ PCR [33]; HPV DNA detection by MY09/11 and GP5+/6+ PCR [14].
¶ Elfgren et al. [14] included a population with baseline HPV DNA infection, age-matched to control population without
baseline HPV DNA infection; analyses in Winer et al. [17] and Koutsky et al. [59] were stratified by baseline HPV DNA
status; analyses in Clavel et al. [30] were stratified by baseline HC2+ status.

women aged <30 years. Follow-up times available (Appendix Table A1)., consistent with the data sum-
for women aged <30 years were also relatively shorter marized in the figures. The summary estimates be-
compared to those for women aged 530 years (over- tween different categories of a variable, however, did
all 16–60 months). In women with baseline HPV not reach statistical significance.
infection, incidence proportion estimates for HSIL/
CIN 2/3 ranged from 1·4% (95% CI 1·0–1·8) to
35·7% (95% CI 29·7–41·7) over follow-up time be- Cumulative incidence of LSIL/CIN 1
tween 24 and 60 months (Fig. 1d ). However, in Estimates of incidence proportion for LSIL/CIN 1
women who were baseline HPV negative or in mixed were stratified by baseline HPV status (HPV positive,
populations, incidence proportion estimates for HPV negative, or mixed population) and further strati-
HSIL/CIN 2/3 were consistently 45% over follow-up fied by age group (<30 years, 530 years) (Fig. 2). In
times of 24–52 months and 16–60 months, respectively women aged 530 years with baseline HPV infection,
(Fig. 1e, f). incidence proportion estimates appeared to increase
In our meta-analysis, the summary incidence pro- with longer follow-up time, although they were ge-
portion estimates of HSIL/CIN 2/3 were consistently nerally <10%, except in women in Amsterdam, The
higher across all variables in women with a baseline Netherlands (median age 37 years; incidence pro-
HPV infection, compared to mixed population portion of CIN 1 at 34 months = 15·9%) [12], as
and those without a baseline HPV infection well as in population-based screening in Sweden
6
J. Ting and others
Table 2(a). Selected characteristics of studies on incident high-grade intraepithelial lesions or cervical intraepithelial neoplasia 2 and 3

Mean/ Mean/median
First-named median follow-up time No. of Outcome Baseline HPV Incidence
author [ref.] Region age (years) (months) women diagnosis infection proportion (%) 95% CI

Age <30 years


Kjaer [23] Denmark 27·9 24 10 177 HSIL Overall 12·38 10·23–14·52
HC2+ 35·70 29·70–41·70
HC2- 3·51 2·10–4·91
Sigurdsson [34] Iceland 20·5 52 282 CIN 2/3 Overall 4·97 2·43–7·50
Woodman [46] UK 18 29 1075 CIN 2/3 HPV DNA- 2·61 1·65–3·56
Cuschieri [45] UK 29·6 37 126 LSIL HPV DNA+ 3·17 0·14–6·24
Brown [16] USA, Europe, Asia 19·8 42 4778 CIN 2/3 HPV DNA+ 1·40 0·73–2·07
Sawaya [55] USA <30 16 12 194 HSIL Overall 0·67 0·53–0·82
Carter [54] USA 19 26 235 HSIL Overall 2·13 0·28–3·97
Winer [17] USA 19·2 36 119 CIN 2/3 HPV DNA+ 11·10 6·50–18·50
Koutsky [59] USA 26 24 241 CIN 2/3 HPV DNA+ 28·00 14·00–41·00
24 HPV DNA- 3·00 0–8·50
Schiffman [58] USA 24 51·6 3408 CIN 2/3 HR-HPV+ 4·66 3·90–5·43
HR-HPV- 1·06 0·14–1·99
Lehtinen [71] USA, Europe, Asia 15–25 48 1983 CIN 2/3 HR-HPV- 3·00 2·23–3·72
Coker [57] USA 29·3 60 2905 HSIL Overall 0·76 0·44–1·1
Harper [70] USA, Brazil 20·7 48 497 CIN 3 HC2- 2·21 0·92–3·51
Age 530 years
Clavel [30] France 41 36 4401 HSIL HC2+ 7·50 5·63–9·64
HC2- 0 –
Bory [31] France 39 4–36 3091 CIN 2/3 Overall 1·42 1·00–1·84
4–10 HC2+ 6·37 4·51–8·24
9–36 HC2- 0·08 0–0·20
Dalstein [29] France 35·7 22 652 CIN 2/3 Overall 3·07 1·74–4·39
Schneider [33] Germany 36·2 12 954 HSIL Overall 4·04 2·74–5·33
Petry [49] Germany 42·7 60 7372 CIN 2/3 HPV DNA- 0 –
Giorgi-Rossi [35] Italy 25–50 54 347 735 CIN2/3 Overall 0·20 0·18–0·21
Bulk [10] The Netherlands 38·5 18 763 CIN 2/3 HR-HPV+ 12·00 9·70–16·00
Hopman [12] The Netherlands 37 34 44 CIN 2/3 HPV DNA+ 18·18 6·78–29·58
Rozendaal [40] The Netherlands 42 40 1276 CIN 3 Overall 0·55 0·14–0·95
Rozendaal [47] The Netherlands 43 76·8 2250 CIN 3 HR-HPV+ 9·92 4·59–15·24
HR-HPV- 0·05 0–0·14
Table 2(a) (cont.)

Mean/ Mean/median
First-named median follow-up time No. of Outcome Baseline HPV Incidence
author [ref.] Region age (years) (months) women diagnosis infection proportion (%) 95% CI

Bulkmans [38] The Netherlands 45 55·2 2687 CIN 2/3 Overall 0·36 0·11–0·60
HR-HPV+ 6·45 0·34–12·57
Bulkmans [36] The Netherlands 41 86 8330 CIN 2/3 Overall 1·11 0·80–1·40
86 CIN 2/3 HPV DNA- 0·40 0·20–0·50
Bos [39] The Netherlands 75–99 87 448 983 CIN 2/3 Overall 0·78 0·73–0·79
Nygard [41] Norway 41·4 24 526 661 CIN 2 Overall 0·20 0·17–0·24
Naucler [11] Sweden 35·1 49 5696 CIN 2/3 Overall 2·60 2·19–3·01
HR-HPV+ 27·50 23·50–31·90
HR-HPV- 0·40 0·25–0·61
Cuschieri [45] UK 37·5 26 225 HSIL HPV DNA- 0 –
Kitchener [48] UK 20–64 36 10 299 CIN 2/3 HC2+ 2·83 1·80–3·87
HC2- 2·61 1·65–3·56
Peto [13] UK 37·5 60 44 611 CIN 3 Overall 9·00 8·73–9·26
Cuzick [44] UK 46 108 2982 CIN 2/3 Overall 2·20 1·63–2·77
Blanks [43] UK 40–49 120 39 841 HSIL Overall 0·28 0·24–0·34
Sawaya [55] USA 30–49 16 53 621 HSIL Overall 0·22 0·20–0·25
Thrall [50] USA 46·0 18 2719 HSIL Overall 0 –
HC2+ 0 –
HC2- 0 –
Schiffman [58] USA 40 120 6220 CIN 2/3 HR-HPV+ 6·00 5·41–6·61
HR-HPV- 1·53 −0·18–3·05

Cervical cancer incidence global review


Sherman [51] USA 35·9 122 20 156 CIN 3 Overall 1·38 1·10–1·67
Gontijo [61] Brazil 32 24 234 CIN 2/3 HC2+ 0·70 0–2·07
HC2- 0 –
Trottier [63] Brazil 32·9 48 1533 HSIL Overall 1·57 0·94–2·19
Rodriguez [26] Costa Rica 41 36 494 CIN 2/3 HPV DNA+ 6·55 4·64–8·47
HPV DNA- 3·85 2·15–5·54
Mitchell [64] Australia 37 36 18 618 CIN2/3 Overall 0·48 0·38–0·58
Shi [66, 72] China 35–45 60 1509 CIN 2/3 Overall 0·80 0·35–1·24
HC2+ 5·65 2·22–9·05
HC2- 0·15 0–0·36
Monteiro [60] Brazil <20 60 403 HSIL Overall 2·98 1·32–4·64

CI, Confidence interval; CIN, cervical intraepithelial neoplasia; HC2, hybrid capture 2; HR-HPV, high-risk human papillomavirus; HSIL, high-grade squamous
intraepithelial lesions.

7
8
J. Ting and others
Table 2(b). Selected characteristics of studies on incident low-grade intraepithelial lesions or cervical intraepithelial neoplasia 1

First-named Mean/ median Follow-up No. of Outcome Baseline HPV Incidence


author [ref.] Region age (years) time (months) women diagnosis infection proportion (%) 95% CI

Age <30 years


Kjaer [23] Denmark 27·9 24 10 177 LSIL Overall 12·71 10·54–14·88
HC2+ 25·30 19·90–30·70
HC2- 7·93 5·86–9·99
Kjaer [15] Denmark 27 127·2 7218 LSIL+ Overall 5·01 4·51–5·52
HC2+ 13·26 11·37–15·16
HC2- 3·32 2·87–3·78
Sigurdsson [34] Iceland 20·5 52 282 CIN 1 Overall 14·54 10·43–18·65
Woodman [46] UK 18 29 1075 LSIL HPV DNA- 9·58 7·82–11·34
Cuschieri [45] UK 29·6 37 126 LSIL HPV DNA+ 3·17 0·11–6·24
Brown [16] USA, Europe, Asia 19·8 42 4778 CIN 1+ HPV DNA+ 4·40 3·75–5·00
Sawaya [55] USA <30 16 12 194 LSIL Overall 3·22 2·91–3·54
Carter [54] USA 19 26 235 LSIL Overall 15·74 11·09–20·40
Winer [17] USA 19·2 36 119 LSIL+ HPV DNA+ 47·20 38·90–56·40
HPV DNA- 1·60 0·70–4·00
Coker [57] USA 29·3 60 2905 LSIL Overall 13·91 12·65–15·26
Moscicki [53] USA 19·7 60 496 LSIL HPV DNA+ 21·00 17·00–25·00
Harper [70] USA, Brazil 20·7 48 497 CIN 1 HC2- 2·62 1·21–4·02
Monteiro [60] Brazil <20 60 403 LSIL Overall 28·04 23·65–32·42
Age 530 years
Clavel [32] France 35 5 204 CIN 1 Overall 6·37 3·02–9·73
HC2+ 7·33 3·16–11·51
HC2- 3·70 −0·13–8·74
Hopman [12] The Netherlands 37 34 44 CIN 1 HPV DNA+ 15·91 5·10–26·71
Rozendaal [40] The Netherlands 42 40 1276 CIN 1 Overall 0·24 −0·03–0·50
Bos [39] The Netherlands 75–99 87 448 983 CIN 1 Overall 0·52 0·50–0·55
Kjaer [15] Denmark 45 120 1305 LSIL+ Overall 3·53 2·52–4·52
HC2+ 21·28 9·58–32·98
HC2- 2·86 1·94–3·78
Elfgren [14] Sweden 35 64·5 90 LSIL+ HPV DNA+ 12·23 3·06–21·42
HPV DNA- 9·75 0·67–18·84
Cuschieri [45] UK 37·5 26 99 LSIL HPV DNA- 1·01 −0·10–2·98
Cuzick [44] UK 46 77 2982 CIN 1 Overall 0·80 0·45–1·14
Blanks [43] UK 40–49 120 39 841 LSIL Overall 0·45 0·38–0·51
Cervical cancer incidence global review 9

(median age 35 years; incidence proportion of LSIL+

CI, Confidence interval; CIN, cervical intraepithelial neoplasia; HC2, hybrid capture 2; HR-HPV, high-risk human papillomavirus; HSIL, high-grade squamous
at 65 months = 12·2%) [14] and in Denmark (median

3·79–14·73
0·84–0·95
0·95–2·24

0·45–1·50
5·20–7·60
0·90–1·30

5·05–7·48
0·46–0·68
age 45 years; incidence proportion of LSIL+ at 120
95% CI

0–4·45
months = 21·3%) [15] (Fig. 2a, Table 2a). In women
aged 530 years who were baseline HPV negative or


in mixed populations, incidence proportion estimates
were generally <5% over 5–120 months, remaining
proportion (%)

relatively constant across follow-up time (Fig. 2b, c).


Incidence

For women aged <30 years, incidence proportion


0·89
1·59
9·26
9·75
6·40
1·10
2·10

6·26
0·57
estimates for LSIL/CIN 1 across follow-up time in
0

women with baseline HPV infection were more vari-


able than that for women aged 530 years, ranging
from 4·4% (95% CI 3·8–5·0) over 42 months [16]
Baseline HPV

to 47·2% (95% CI 38·9–56·4) at 36 months [17]


infection

(Fig. 2d). In women without baseline HPV infection,


Overall
Overall

Overall
Overall
HC2+

HC2+

HC2+
HC2-

HC2-

HC2-

incidence proportion estimates were consistently <10%


over follow-up time (24–127 months) (Fig. 2e). In
mixed populations, however, incidence proportion
diagnosis

estimates for LSIL/CIN 1 appeared to increase with


Outcome

LSIL+

CIN 1

CIN 1

longer follow-up time, from 3·2% (95% CI 2·9–3·5)


LSIL
LSIL

LSIL

at 16 months to 28·0% (95% CI 23·7–32·4) at


60 months (Fig. 2f ).
women
No. of

2719

1533
53 621

16 626

234

18 618

D I S C US S I O N
This systematic review included data on over
time (months)

2.2 million women from 54 studies to summarize the


Follow-up

incidence of cervical lesions in women with normal


baseline cervical cytology. Incidence of both high-
grade (HSIL/CIN 2/3) and of low-grade (LSIL/CIN
16
18

57

24

48
36

1) lesions appeared to increase with longer study


follow-up in women aged 530 years with a baseline
Mean/ median

HPV infection compared to in women negative for


age (years)

HPV at baseline or in HPV-mixed populations


within whom incidence remained relatively low. In
30–49

women aged 530 years with baseline HPV infection,


46·0

35·9

32·9
32

37

the incidence of HSIL/CIN 2/3 did not appear to be


notably higher than that of LSIL/CIN 1 in the same
age group, and also did not appear higher than
in young women aged <30 years with baseline HPV
infection.
Australia

Our finding of a higher risk of HSIL/CIN 2/3 with


Region

Brazil

Brazil

longer follow-up time in women aged 530 years


USA
USA

USA

with prevalent HPV infection, but not in women


intraepithelial lesions.
Table 2(b) (cont.)

aged <30 years or in women without HPV at baseline,


supports the current U.S. Prevention Services Task
Mitchell [64]

Force (USPSTF) recommendation for HPV co-testing


First-named

Gontijo [61]

Trottier [63]
author [ref.]

Sawaya [55]
Thrall [50]

Castle [52]

with cytology of women aged 530 years [6]. In con-


trast, HPV infections detected in younger women at
screening may represent newly acquired infections,
10
J. Ting and others
(a) (b) (c)

(d) (e) (f )

Fig. 1. Normal baseline cervical cytology to HSIL/CIN2/3 by age and HPV infection at baseline. CIN, Cervical intraepithelial neoplasia; HPV, human papillomavirus;
mixed HPV, studies that include both HPV-positive and HPV-negative women; HSIL, high-grade squamous intraepithelial lesions.
(a) (b) (c)

(d) (e) (f )

Cervical cancer incidence global review


Fig. 2. Normal baseline cervical cytology to incident LSIL/CIN 1 by age and HPV infection at baseline. CIN, Cervical intraepithelial neoplasia; HPV, human
papillomavirus; mixed HPV, studies that include both HPV-positive and HPV-negative women; LSIL, low-grade squamous intraepithelial lesions.

11
12 J. Ting and others

the majority of which may be controlled and cleared to 216 months [82]. Incidence proportion for HSIL/
and thus rarely lead to CIN 2/3 or cancer [16, 18–20]. CIN 2/3 in women aged 530 years with baseline HPV
The time from initial HPV infection to HSIL/CIN ranged from 0·7% (95% CI 0·14–1·27) over 36 months
2/3 can be short in some young women, consistent [79] to 7·0% (95% CI 5·7–8·6) over 216 months [82],
with results from the control arms of prophylactic vac- and that for women without baseline HPV from 0%
cine trials [21, 22]. In female university students aged over 60 months [75] to 1·4% (95% CI 1·2–1·7) over
18–20 years in Seattle, USA, the overall incidence 216 months [82], consistent with our current results.
of CIN 2/3 was 11·1% at 36 months, and that of However, in studies comparing women aged <30
HPV-16- and HPV-18-associated CIN 2/3 was 27% and 530 years [77, 79], HSIL/CIN 2/3 incidence
at 36 months [17]. Similarly, another study in proportion overall appeared somewhat higher in
Denmark found a high incidence of HSIL (35·7% at women aged <30 years than in women aged 530
24 months) in young women from the general popu- years (3·8% vs. 1·3% over 126 months [77], and 0·4%
lation (mean age 25 years) [23]. Although data on vs. 0·1% over 36 months [79]).
CIN 2/3 regression in these university students are In the USA, prophylactic HPV vaccines are
not generally available given that cytological abnor- licensed and recommended for women aged 426
malities were referred to colposcopy and treatment, years [27, 28]. There is currently a lack of data sug-
if indicated [17], another study showed that most gesting high efficacy of vaccination in preventing
CIN 2 cases (∼70%) in young women (mean age infections or cervical lesions in women aged >26
20 years) regressed within 3 years [24]. Although our years. Previous data on women aged 24–45 years
findings of non-negligible HSIL/CIN 2/3 incidence in showed that the quadrivalent HPV vaccine was effica-
women aged <30 years with prevalent HPV infection cious in women with no evidence of previous exposure
suggest potential benefit of HR-HPV DNA co-testing with the vaccine-types (defined by baseline sero-
with Pap for some women aged <30 years, data on re- negativity for HPV-6/-11/-16/-18 and DNA negative
gression of HSIL/CIN 2/3 in young women are needed by PCR in cervical specimens), but not in women
to determine the clinical significance of these lesions, with prevalent HPV infections [8]. However, this clini-
particularly in terms of their potential to progress cal trial had a relatively short follow-up time (mean
to cervical cancer [6]. 2·2 years), and used an endpoint of combined incident
We found that in women aged 530 years with HPV infection and cervical disease [8]. Further
baseline HPV infection, the incidence of HSIL/CIN vaccine efficacy data on incident HPV infection and
2/3 over follow-up time did not appear to be notably subsequent HSIL/CIN 2/3 development or regression
higher than that of LSIL/CIN 1. It is unclear from in women aged >26 years with longer follow-up
this literature review if these incident LSIL/CIN time will determine if relatively older women can po-
1 cases were due to recently acquired or persistent tentially benefit from HPV vaccination, particularly
HPV infections. Recent data have also found that those in whom prevalent HPV infection and cervical
most (85%) incident HPV infections in older women disease are no longer present.
were detected during periods of sexual abstinence or It is important to note that incidence data of
monogamy, suggesting that the risk of newly detected cervical lesions in women aged <30 years are generally
HPV infections in older women are likely due to re- limited (n = 18), compared to those in women aged
activation of latent infections [25]. Previous longitudi- 530 years (n = 38). Data on follow-up time was also
nal data from Costa Rica showed that new HPV generally shorter for women aged <30 years than for
infections in older women (542 years) rarely progress women aged 530 years. The lack of data on women
to CIN 2/3 or cancer within 3 years, but longer follow- aged <30 years limited our ability to draw definitive
up of women is needed to understand their true risk conclusions about the acquisition risk of cervical
of CIN2/3 development [26]. lesions over time, as well as about the benefits of
We further identified nine eligible studies published HPV testing for cervical cancer screening in this
between 30 September 2012 and 28 February 2014 younger age group. Further, we are unable to calcu-
[75–83], of which four were from Europe [75–78], late and formally compare incidence rates, stratified
four from the USA [79–82], and one from Australia by covariates such as prevalent HPV status and age,
[83]. Six were population-based studies [75–79, 83], given that few studies reported person-time data. In
and three were screening-based studies [80–82]. studies where we could not determine the baseline
Total follow-up time ranged from 36 months [79] HPV prevalence, we assumed HPV prevalence to be
Cervical cancer incidence global review 13

between >0% and <100% and thus included these proportion of HSIL/CIN2/3 of ∼1% at 24 months.
studies in the ‘mixed’ HPV prevalence category. Consequently, the undefined standard errors were esti-
Data on the ‘mixed’ group were included so that we mated by adding 0·01 to the number of women with
may observe the overall pattern of incident lesions incident HSIL/CIN 2/3 and 0·09 to the number of
over follow-up time in a given population, stratified women without incident HSIL/CIN 2/3. Otherwise,
by age. We also did not include data on the incidence if the standard error was not reported and could
of HPV type-specific associated lesions, given that not be calculated from the information reported in
studies with HPV genotype-specific results were few, the article, the incidence result was not included
and our study was initiated before the publication in the meta-regression analyses.
of the revised USPSTF screening guidelines in 2012
recommending HPV genotype-specific testing for
Meta-regression analyses
HPV-16 or HPV-16/-18 for women who co-tested
HPV positive and cytologically negative [5, 6]. Meta-regression was used to produce summary esti-
In conclusion, this systematic review showed that mates of the proportion of women who developed in-
the incidence of HSIL/CIN 2/3 in women aged 530 cident cervical abnormalities. When results from the
years is higher with prevalent baseline HPV infection, same study population were included in different
and increased with longer follow-up time. We found strata of an analysis, an indicator variable for that
that some young women aged <30 years with pre- study population was included in the meta-regression
valent HPV infection are at risk of developing model to account for the lack of independence.
HSIL/CIN 2/3 within a relatively short period of Meta-regression analysis was conducted in Stata
time, although we did not abstract data on the re- version 10 (StataCorp, USA).
gression of these lesions to determine if they would To formally compare incidence proportion across
be likely to advance to cancer. Our finding of a non- studies and by key study characteristics, random-
negligible increase in LSIL/CIN 1 over follow-up in effects meta-regression was used, with the among-
women aged 530 years who had a baseline HPV study variance estimated by restricted maximum like-
infection may indicate an HPV infection that is po- lihood [74]. Stratified summary estimates allowed de-
tentially progressing to higher grade cervical lesions. scriptive comparisons across individual categories of
However, data on the potential of these LSIL/CIN 1 study characteristics by providing separate summary
to progress to more severe cervical disease should be estimates and 95% confidence intervals for each cate-
summarized to help inform cervical cancer prevention gory. Studies were allowed to contribute to more
strategies. than one category to reduce any potential influence
of the decision rules on the distribution of study and
population characteristics.

APPENDIX
Meta-regression analyses AC KN OWL ED GE MEN T S
Standardization of estimates of incidence of cervical This study was funded by an unrestricted grant by
lesions GlaxoSmithKlineVaccines, Global Vaccine Develop-
ment, Wavre, Belgium.
In the meta-regression analyses, length of follow-up
was centred at 24 months. In cases where the incidence
proportion of abnormal cytology was 0% or 100% and
the standard error was undefined, the following ad- D E C L A RATI O N O F I NT E R E S T
justment was made to calculate the standard error: Jennifer S. Smith has received research grants,
the overall meta-regression model was first fitted with- served on advisory boards, and/or been a speaker
out study characteristics (intercept-only model) using for GlaxoSmithKline, Merck Corporation, Hologic
all studies that did have a defined standard error Gen-Probe, Qiagen, and BD Diagnostics. Sylvia
[73]. This model produced a summary incidence M. Taylor is an employee of GlaxoSmithKline.
14
J. Ting and others
Appendix Table A1. Meta-regression of the incidence proportion of HSIL/CIN 2/3 in women with normal cervical diagnoses at baseline, stratified by baseline
HPV infection status

Baseline HPV-negative (N = 18) Baseline HPV-positive (N = 18) Baseline mixed HPV (N = 24)

Summary incidence Summary incidence Summary incidence


No. proportion* (95% CI) No. proportion† (95% CI) No. proportion† (95% CI)

Study region
Europe 10 0·2 (−0·3 to 0·8) 11 9·6 (5·2 to 13·9) 15 2·2 (0·8 to 3·6)
North America 3 0 (−1·2 to 1·1) 5 2·0 (−6·7 to 10·7) 5 0·9 (−1·3 to 3·1)
Central and South America 2 0·5 (−0·8 to 1·8) 1 0·7 (−12·7 to 14·1) 2 2·5 (−1·0 to 6·0)
Asia, Australia, multi-region 3 0·5 (−0·5 to 1·5) 1 5·1 (0·3 to 19·5) 2 0·9 (−2·4 to 4·2)
Study populations
Young women (15–21 years) 2 0·4 (−0·2 to 1·0) 1 7·0 (−6·7 to 20·7) 2 2·8 (−0·8 to 6·3)
Population-based 7 0·2 (−0·1 to 0·5) 7 11·4 (6·1 to 16·7) 12 1·9 (0·1 to 3·8)
Screening-based†/GYN/family- 9 0·1 (−0·2 to 0·3) 10 4·7 (−0·1 to 9·4) 10 1·7 (0·1 to 3·2)
planning/hospital-based
Mean/median age of women
<30 years 3 0·8 (0·2 to 1·4) 1 7·0 (−8·2 to 22·2) 4 2·8 (0·3 to 5·3)
530 years 15 0·1 (−0·1 to 0·3) 17 7·7 (3·5 to 12·0) 20 1·7 (0·4 to 2·9)
Mean/median length of follow-up
424 months 5 0 (−0·2 to 0·3) 8 6·4 (1·3 to 11·4) 8 1·6 (0 to 3·2)
>24 months 13 0·2 (0·1 to 0·4) 10 7·5 (2·8 to 12·2) 16 1·6 (0·5 to 2·7)
Baseline HPV serostatus
HPV-16/HPV-18 seronegative 2 0·5 (−0·2 to 1·1) 0 − 1 2·1 (−2·7 to 7·0)
No HPV antibody detection 16 0·1 (−0·1 to 1·4) 18 7·7 (3·7 to 11·6) 23 1·8 (0·6 to 3·1)

CI, Confidence interval; CIN, cervical intraepithelial neoplasia; CUI, percent cumulative incidence; GYN, gynecological clinic; HPV, human papillomavirus; HR, high risk;
HSIL, high-grade squamous intraepithelial lesions; LR, low risk.
* Estimates are adjusted for follow-up duration (centred at 24 months), and for study populations that contributed to more than one stratum.
† Includes women at community-based clinics screened through the National Breast and Cervical Cancer Early Detection Program (NBCCEDP) [55].
Cervical cancer incidence global review 15

R E FE RE NC E S 16. Brown DR, et al. The impact of quadrivalent human


papillomavirus (HPV; types 6, 11, 16, and 18) L1 virus-
1. Walboomers JM, et al. Human papillomavirus is a like particle vaccine on infection and disease due to
necessary cause of invasive cervical cancer worldwide. oncogenic nonvaccine HPV types in sexually active
Journal of Pathology 1999; 89: 12–19. women aged 16–26 years. Journal of Infectious Disease
2. Smith JS, et al. Human papillomavirus type distribution 2009; 199: 936–944.
in invasive cervical cancer and high-grade cervical 17. Winer RL, et al. Development and duration of human
lesions: a meta-analysis update. International Journal papillomavirus lesions, after initial infection. Journal
of Cancer 2007; 121: 621–132. of Infectious Disease 2005; 191: 731–738.
3. Moscicki A, et al. Updating the natural history of HPV 18. Wright TC. Natural history of HPV infections. Journal
and anogenital cancer. Vaccine 2006; 24: S42–51. of Family Practice 2009; 59: S3–7.
4. Cuzick J, et al. Overview of the European and North 19. Moscicki AB, et al. Regression of low-grade squamous
American studies on HPV testing in primary cervical intra-epithelial lesions in young women. Lancet 2004;
cancer screening. International Journal of Cancer 2006; 364: 1678–1683.
119: 1095–1101. 20. Rositch AF, et al. Patterns of persistent genital human
5. Saslow D, et al. American Cancer Society, American papillomavirus infection among women worldwide: a
Society for Colposcopy and Cervical Pathology, and literature review and meta‐analysis. International
American Society for Clinical Pathology screening Journal of Cancer 2013; 133: 1271–1285
guidelines for the prevention and early detection of cer- 21. Paavonen J, et al. Efficacy of a prophylactic adjuvanted
vical cancer. American Journal of Clinical Pathology bivalent L1 virus-like-particle vaccine against infection
2012; 137: 516–542. with human papillomavirus types 16 and 18 in young
6. Moyer VA, on behalf of the U.S Preventive women: an interim analysis of a phase III double-blind,
Services Task Force. Screening for cervical cancer: US randomised controlled trial. Lancet 2007; 369: 2161–
Preventive Services Task Force recommendation state- 2170.
ment. Annals of Internal Medicine 2012. 22. Garland SM, et al. Quadrivalent vaccine against
7. Schiffman M, et al. Human papillomavirus and cervical human papillomavirus to prevent anogenital diseases.
cancer. Lancet 2007; 370: 890–907. New England Journal of Medicine 2007; 356: 1928–1943.
8. Muñoz N, et al. Safety, immunogenicity, and efficacy of 23. Kjaer SK, et al. Type specific persistence of high risk
quadrivalent human papillomavirus (types 6, 11, 16, 18) human papillomavirus (HPV) as indicator of high
recombinant vaccine in women aged 24–45 years: a rando- grade cervical squamous intraepithelial lesions in
mised, double-blind trial. Lancet 2009; 373: 1949–1957. young women: population based prospective follow up
9. Skinner SR, et al. HPV-16/18 AS04-adjuvanted vaccine study. British Medical Journal 2002; 325: 572–578.
efficacy in 526-year-old women after 4-year follow-up. 24. Moscicki A, et al. Rate of and risks for regression of
Presented at the International Papillomavirus Con- CIN-2 in adolescents and young women. Obstetrics &
ference, Berlin, 2011. Gynecology 2010; 116: 1373–1380.
10. Bulk S, et al. Risk of high‐grade cervical intra‐epithelial 25. Rositch AF, et al. Contributions of recent and past sex-
neoplasia based on cytology and high‐risk HPV testing ual partnerships on incident human papillomavirus de-
at baseline and at 6‐months. International Journal of tection: acquisition and reactivation in older women.
Cancer 2007; 121: 361–367. Cancer Research 2012; 72: 6183–6190.
11. Naucler P, et al. HPV type-specific risks of high-grade 26. Rodriguez AC, et al. Longitudinal study of human
CIN during 4 years of follow-up: a population-based papillomavirus persistence and cervical intraepithelial
prospective study. British Journal of Cancer 2007; 97: neoplasia grade 2/3: Critical role of duration of infec-
129–32. tion. Journal of the National Cancer Institute 2010;
12. Hopman EH, et al. High risk human papillomavirus 102: 315–324.
in women with normal cervical cytology prior to the de- 27. Markowitz LE, et al. Quadrivalent human papilloma-
velopment of abnormal cytology and colposcopy. virus vaccine: recommendations of the Advisory Com-
British Journal of Obstetrics and Gynaecology 2000; mittee on Immunization Practices (ACIP). Morbidity
107: 600–604. and Mortality Weekly Report 2007; 56(RR02): 1–24.
13. Peto J, et al. Cervical HPV infection and neoplasia 28. Centers for Disease Control, Prevention (CDC). FDA
in a large population-based prospective study: the licensure of bivalent human papillomavirus vaccine
Manchester cohort. British Journal of Cancer 2004; 91: (HPV2, cervarix) for use in females and updated HPV
942–953. vaccination recommendations from the advisory com-
14. Elfgren K, et al. A population-based five-year follow-up mittee on immunization practices (ACIP). Morbidity
study of cervical human papillomavirus infection. and Mortality Weekly Report 2010; 59: 626–629.
American Journal of Obstetrics and Gynecology 2000; 29. Dalstein V, et al. Persistence and load of high‐risk
183: 561–567. HPV are predictors for development of high‐grade
15. Kjaer S, et al. The absolute risk of cervical abnor- cervical lesions: a longitudinal French cohort study.
malities in high-risk human papillomavirus-positive, International Journal of Cancer. 2003; 106: 396–403.
cytologically normal women over a 10-year period. 30. Clavel C, et al. Negative human papillomavirus testing
Cancer Research 2006; 66: 10630–10636. in normal smears selects a population at low risk for
16 J. Ting and others

developing high-grade cervical lesions. British Journal of former soviet union. Journal of Clinical Microbiology
Cancer. 2004; 90: 1803–1808. 2004; 42: 505–511.
31. Bory JP, et al. Recurrent human papillomavirus infec- 43. Blanks RG, et al. Risk of cervical abnormality after age
tion detected with the hybrid capture II assay selects 50 in women with previously negative smears. British
women with normal cervical smears at risk for develop- Journal of Cancer 2009; 100: 1832–1836.
ing high grade cervical lesions: a longitudinal study 44. Cuzick J, et al. Long‐term follow‐up of cervical abnor-
of 3,091 women. International Journal of Cancer. malities among women screened by HPV testing and
2002; 102: 519–525. cytology-results from the Hammersmith study.
32. Clavel C, et al. Human papillomavirus detection by the International Journal of Cancer 2008; 122: 2294–2300.
hybrid capture II assay: a reliable test to select women 45. Cuschieri KS, et al. Persistent high risk HPV infection
with normal cervical smears at risk for developing cervi- associated with development of cervical neoplasia in
cal lesions. Diagnostic Molecular Pathology 2000; 9: a prospective population study. Journal of Clinical
145–150. Pathology 2005; 58: 946–950.
33. Schneider A, et al. Different detectability of high-risk 46. Woodman CB, et al. Natural history of cervical human
HPV in smears from incident and prevalent high-grade papillomavirus infection in young women: A longitudi-
squamous intraepithelial lesions of the cervix. Gyneco- nal cohort study. Lancet 2001; 357: 1831–1836.
logic Oncology 1997; 65: 399–404. 47. Rozendaal L, et al. PCR based high risk HPV testing is
34. Sigurdsson K, et al. The efficacy of HPV 16/18 vaccines superior to neural network based screening for predict-
on sexually active 18–23 year old women and the impact ing incident CIN III in women with normal cytology
of HPV vaccination on organized cervical cancer screen- and borderline changes. Journal of Clinical Pathology
ing. Acta Obstetrica et Gynecologica Scandinavica 2009; 2000; 53: 606.
88: 27–35. 48. Kitchener HC, et al. HPV testing in combination with
35. Giorgi-Rossi P, Baiocchi D, Ciatto S, Endocervical Cell liquid-based cytology in primary cervical screening
Italian Working Group of Cervical Cancer Screening (ARTISTIC): a randomised controlled trial. Lancet
Group. Risk of CIN 2 in women with a pap test without Oncology 2009; 10: 672–682.
endocervical cells vs. those with a negative pap test with 49. Petry KU, et al. Inclusion of HPV testing in routine
endocervical cells: a cohort study with 4·5 years of cervical cancer screening for women above 29 years in
follow-up. Acta Cytologica 2010; 54: 265–271. Germany: results for 8,466 patients. British Journal of
36. Bulkmans NWJ, et al. Human papillomavirus DNA Cancer 2003; 88: 1570–1577.
testing for the detection of cervical intraepithelial neo- 50. Thrall M, et al. High-risk HPV testing in women
plasia grade 3 and cancer: 5-year follow-up of a rando- 30 years or older with negative papanicolaou tests.
mised controlled implementation trial. Lancet 2007; American Journal of Clinical Pathology 2010; 133:
370: 1764–1772. 894–898.
37. Berkhof J, et al. Human papillomavirus type-specific 51. Sherman ME, et al. Baseline cytology, human papillo-
18-month risk of high-grade cervical intraepithelial mavirus testing, and risk for cervical neoplasia: a
neoplasia in women with a normal or borderline/mildly 10-year cohort analysis. Journal of the National
dyskaryotic smear. Cancer Epidemiology, Biomarkers & Cancer Institute 2003; 95: 46–52.
Prevention 2006; 15: 1268–1273. 52. Castle PE, et al. Absolute risk of a subsequent abnor-
38. Bulkmans NWJ, et al. Long-term protective effect of mal pap among oncogenic human papillomavirus
high-risk human papillomavirus testing in population- DNA‐positive, cytologically negative women. Cancer
based cervical screening. British Journal of Cancer 2002; 95: 2145–2151.
2005; 92: 1800–1802. 53. Moscicki AB, et al. Risks for incident human papillo-
39. Bos AB, et al. Endocervical status is not predictive of mavirus infection and low-grade squamous intraepithe-
the incidence of cervical cancer in the years after nega- lial lesion development in young females. Journal of the
tive smears. American Journal of Clinical Pathology American Medical Association 2001; 285: 2995–3002.
2001; 115: 851–855. 54. Carter JJ, et al. The natural history of human papillo-
40. Rozendaal L, et al. PCR-based high-risk HPV test in mavirus type 16 capsid antibodies among a cohort of
cervical cancer screening gives objective risk assessment university women. Journal of Infectious Disease 1996;
of women with cytomorphologically normal cervical 174: 927–936.
smears. International Journal of Cancer 1996; 68: 55. Sawaya GF, et al. Frequency of cervical smear abnor-
766–769. malities within 3 years of normal cytology. Obstetrics
41. Nygard JF, et al. CIN 2/3 and cervical cancer in an & Gynecology 2000; 96: 219–223.
organised screening programme after an unsatisfactory 56. Ho GYF, et al. Natural history of cervicovaginal papil-
or a normal pap smear: a seven-year prospective lomavirus infection in young women. New England
study of the norwegian population-based screening Journal of Medicine 1998; 338: 423–438.
programme. Journal of Medical Screening 2004; 11: 57. Coker AL, et al. High-risk HPVs and risk of cervical
70–76. neoplasia: a nested case-control study. Experimental
42. Syrjanen S, et al. Acquisition of high-risk human papil- and Molecular Pathology 2001; 70: 90–95.
lomavirus infections and pap smear abnormalities 58. Schiffman M, et al. A long-term prospective study of
among women in the new independent states of the type-specific human papillomavirus infection and risk
Cervical cancer incidence global review 17

of cervical neoplasia among 20,000 women in the cervical intraepithelial neoplasia: 4-year end-of-study
Portland Kaiser cohort study. Cancer Epidemiology, analysis of the randomised, double-blind PATRICIA
Biomarkers & Prevention 2011; 20: 1398–1409. trial. Lancet Oncology 2012; 13: 89–99.
59. Koutsky LA, et al. A cohort study of the risk of cervical 72. Shi JF, et al. Estimation of the costs of cervical cancer
intraepithelial neoplasia grade 2 or 3 in relation to papil- screening, diagnosis and treatment in rural Shanxi prov-
lomavirus infection. New England Journal of Medicine ince, China: a micro-costing study. BMC Health
1992; 327: 1272–1278. Services Research 2012; 12: 123.
60. Monteiro DLM, et al. Incidence of cervical intraepithe- 73. Sweeting MJ, Sutton AJ, Lambert PC. What to add to
lial lesions in a population of adolescents treated in pub- nothing? use and avoidance of continuity corrections in
lic health services in Rio De Janeiro, Brazil. Cadernos de meta‐analysis of sparse data. Statistics in Medicine
Saude Publica 2009; 25: 1113–1122. 2004; 23: 1351–1375.
61. Gontijo RC, et al. Human papillomavirus (HPV) infec- 74. Thompson SG, Sharp SJ. Explaining heterogeneity
tions as risk factors for cytological and histological in meta‐analysis: a comparison of methods. Statistics
abnormalities in baseline PAP smear-negative women in Medicine 1999; 18: 2693–2708.
followed-up for 2 years in the LAMS study. European 75. Luyten A, et al. Early detection of CIN3 and cervical
Journal of Obstetrics & Gynecology and Reproductive cancer during long-term follow-up using HPV/Pap
Biology 2007; 133: 239–246. smear in co-testing and risk-adapted follow-up in a
62. Schlecht NF, et al. Persistent human papillomavirus locally organized screening programme. International
infection as a predictor of cervical intraepithelial neo- Journal of Cancer. Published online: 9 August 2013.
plasia. Journal of the American Medical Association doi:10.1186/1471-2407-13-379.
2001; 286: 3106–3114. 76. Petry KU, et al. Annual Papanicolaou screening for 5
63. Trottier H, et al. Human papillomavirus infections with years among human papillomavirus-negative women.
multiple types and risk of cervical neoplasia. Cancer BMC Cancer 2013; 13: 379
Epidemiology, Biomarkers & Prevention 2006; 15: 1274. 77. Thomsen LT, et al. High-risk and low-risk human papil-
64. Mitchell HS. Longitudinal analysis of histologic high‐ lomavirus and the absolute risk of cervical intraepithe-
grade disease after negative cervical cytology according lial neoplasia and cancer. Obstetrics & Gynecolocy
to endocervical status. Cancer Cytopathology 2001; 93: 2014; 123: 57–64
237–240. 78. Elfstrom KM, et al. Long term duration of protective
65. Mitchell H, Medley G. Longitudinal study of women effect for HPV negative women: follow-up of primary
with negative cervical smears according to endocervical HPV screening randomized controlled trial. British
status. Lancet 1991; 337: 265–267. Medical Journal 2014; 138: g130
66. Shi JF, et al. Human papillomavirus testing for cervical 79. Wheeler CM, et al. The influence of type-specific human
cancer screening: results from a 6-year prospective study papillomavirus infections on the detection of cervical
in rural china. America Journal of Epidemiology 2009; precancer and cancer: a population-based study of op-
170: 708–716. portunistic cervical screening in the United States.
67. The FUTURE II study group. Quadrivalent vaccine International Journal of Cancer. Published online:
against human papillomavirus to prevent high-grade 15 April 2014. doi:10.1002/ijc.28605.
cervical lesions. New England Journal of Medicine 80. Katki HA, et al. Five-year risk of CIN 3+ to guide the
2007; 356: 1915–1927. management of women aged 21 to 24 years. Journal
68. The FUTURE II study group. Effect of prophylactic of Lower Genital Tract Disease 2013; 5 (Suppl. 1):
human papillomavirus L1 virus-like-particle vaccine S64–68
on risk of cervical intraepithelial neoplasia grade 2, 81. Katki HA, et al. Five-year risk of CIN 3+ and cervical
grade 3, and adenocarcinoma in situ: a combined analy- cancer among women who test Pap-negative but are
sis of four randomised clinical trials. Lancet 2007; 369: HPV-positive. Journal of Lower Genital Tract Disease
1861–1868. 2013; 5 (Suppl. 1): S56–63
69. Tay EH, et al. Clinical trial experience with prophy- 82. Castle PE, et al. Clinical human papillomavirus de-
lactic HPV 6/11/16/18 VLP vaccine in young women tection forecasts cervical cancer risk in women over 18
from the Asia-Pacific region. International Journal of years of follow-up. Journal of Clinical Oncology 2012;
Gynecology & Obstetrics 2008; 102: 275–283. 30: 3044–3050
70. Harper DM, et al. Sustained efficacy up to 4·5 years of 83. Sultana F, et al. High-grade cervical abnormalities and
a bivalent L1 virus-like particle vaccine against human cervical cancer in women following a negative Pap
papillomavirus types 16 and 18: Follow-up from a ran- smear with and without an endocervical component: a
domised control trial. Lancet 2006; 367: 1247–1255. cohort study with 10 years of follow-up. International
71. Lehtinen M, et al. Overall efficacy of HPV-16/18 Journal of Cancer. Published online: 1 February 2014.
AS04-adjuvanted vaccine against grade 3 or greater doi:10.1002/ijc.28756.

View publication stats